CN104469848A - A Method of Load Balancing Between Heterogeneous Access Networks - Google Patents

Abstract

Translated fromChinese

本发明提供一种异构接入网络之间负载均衡的方法，该方法包括：计算所述异构接入网络中各接入网络的用户接入效用，其中用户接入效用用于衡量接入网络的负载；根据所述用户接入效用计算公平性指数；以及，根据所述公平性指数判断是否进行负载平衡。本发明减少了负载均衡过程中的信令交互次数，降低了信令负荷，并简化了用户接入网络选择过程。

The present invention provides a method for load balancing between heterogeneous access networks, the method comprising: calculating the user access utility of each access network in the heterogeneous access network, wherein the user access utility is used to measure access network load; calculating a fairness index according to the user access utility; and judging whether to perform load balancing according to the fairness index. The invention reduces the number of signaling interactions in the load balancing process, reduces the signaling load, and simplifies the user access network selection process.

Description

Translated fromChinese

一种异构接入网络之间负载均衡的方法A Method of Load Balancing Between Heterogeneous Access Networks

技术领域technical field

本发明涉及宽带移动通信技术领域，以及更具体地，涉及一种异构接入网络之间负载均衡的方法。The present invention relates to the technical field of broadband mobile communication, and more specifically, relates to a load balancing method among heterogeneous access networks.

背景技术Background technique

随着移动通信技术的快速发展，无线移动通信系统呈现异构多元化。目前，无线网络中典型的无线接入技术包含2G/3G/4G、WiFi、UWB(超宽带Ultra Wideband)、Buletooth(蓝牙)等，这些接入技术在覆盖范围、移动性和业务特性等方面有着很大差异，适用的场景各有侧重，彼此之间难以相互取代，因此异构无线网络间的融合是无线通信行业的发展趋势。当用户处于多个无线接入网络共同覆盖的范围时，需要根据一定的策略选择一种接入网络。然而，无线通信中由于用户的移动性经常存在部分热点区域流量需求较大，甚至出现网络拥堵、呼叫阻塞，然而周边的非热点区域业务量较低、网络资源空闲。这样无线资源不能得到合理利用，降低了网络的整体性能。异构网络负载均衡就是针对这种情况而提出的，通过异构网络间的负载均衡，可以有效地提高无线资源利用率、提升网络容量，避免局部网络拥塞。With the rapid development of mobile communication technology, wireless mobile communication systems are heterogeneous and diversified. At present, typical wireless access technologies in wireless networks include 2G/3G/4G, WiFi, UWB (Ultra Wideband), and Bluetooth (Bluetooth), etc. These access technologies have advantages in terms of coverage, mobility, and service characteristics. They are very different, and the applicable scenarios have different emphases, and it is difficult to replace each other. Therefore, the integration of heterogeneous wireless networks is the development trend of the wireless communication industry. When a user is in a range covered by multiple wireless access networks, it is necessary to select an access network according to a certain strategy. However, due to the mobility of users in wireless communication, some hotspot areas often have high traffic demand, and even network congestion and call blocking, while surrounding non-hotspot areas have low traffic and idle network resources. In this way, radio resources cannot be used reasonably, which reduces the overall performance of the network. Heterogeneous network load balancing is proposed for this situation. Through load balancing among heterogeneous networks, the utilization rate of wireless resources can be effectively improved, network capacity can be increased, and local network congestion can be avoided.

现有的异构接入网络之间的负载均衡方法主要包括：基于进化博弈的方法、基于用户和网络相互评价的方法，以及基于系统吞吐量和用户公平性的方法。然而，第一类方法在重复迭代进化过程中需要接入网络和中心控制器进行多次数据交互，信令负荷高且算法复杂度高。后两类方法中，确定用户选择接入网络的方法需要遍历各个网络，复杂度较高，并不适用于实际异构网中的网络间负载均衡。Existing load balancing methods between heterogeneous access networks mainly include methods based on evolutionary games, methods based on mutual evaluation between users and networks, and methods based on system throughput and user fairness. However, the first type of method requires multiple data interactions between the access network and the central controller during the iterative evolution process, resulting in high signaling load and high algorithm complexity. Among the latter two types of methods, the method of determining the user's choice of accessing the network needs to traverse each network, which has high complexity and is not suitable for load balancing between networks in an actual heterogeneous network.

可见，尽管上述现有技术针对接入网络间负载失衡的情况，均给出了有效的机制，使网络间负载达到均衡。但按照上述现有技术，只要存在提高系统性能的可能，就要执行负载均衡算法，而均衡的过程需要大量的数据交互，且复杂度高。It can be seen that although the above-mentioned prior art provides an effective mechanism for load imbalance between access networks, so as to balance the load among the networks. However, according to the above prior art, as long as there is a possibility of improving system performance, a load balancing algorithm must be executed, and the balancing process requires a large amount of data interaction and is highly complex.

发明内容Contents of the invention

本发明针对上述现有技术存在的问题，提出一种异构接入网络之间负载均衡的方法，包括：Aiming at the problems existing in the above-mentioned prior art, the present invention proposes a load balancing method between heterogeneous access networks, including:

步骤1)、计算所述异构接入网络中各接入网络的用户接入效用；其中，用户接入效用用于衡量接入网络的负载；Step 1), calculating the user access utility of each access network in the heterogeneous access network; wherein, the user access utility is used to measure the load of the access network;

步骤2)、根据所述用户接入效用计算公平性指数；Step 2), calculating a fairness index according to the user access utility;

步骤3)、根据所述公平性指数判断是否进行负载平衡。Step 3), judging whether to perform load balancing according to the fairness index.

上述方法中，步骤1)包括：各接入网络统计接入用户数并计算用户接入效用。其中，所述用户接入效用是接入网络中用户的平均吞吐量。In the above method, step 1) includes: each access network counts the number of access users and calculates user access utility. Wherein, the user access utility is the average throughput of users in the access network.

上述方法中，步骤2)包括：In the above-mentioned method, step 2) comprises:

计算不同接入网络之间用户接入效用的Jain’s公平性指数，其中，根据下式计算该公平性指数LFI：Calculate the Jain's fairness index of user access utility between different access networks, where the fairness index LFI is calculated according to the following formula:

$\mathrm{<span><span>LFI</span>LFI</span>}<span><span>=</span>=</span>{<span><span>(</span>(</span>\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>)</span>)</span>}^{\mathrm{<span><span>2</span>2</span>}}<span><span>/</span>/</span><span><span>(</span>(</span>\mathrm{<span><span>N</span>N</span>}\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}<span><span>(</span>(</span>{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}^{\mathrm{<span><span>2</span>2</span>}}<span><span>)</span>)</span><span><span>)</span>)</span>$

其中，S表示所有接入网络的集合，N表示接入网络的总数，Π_i表示接入网络i的用户接入效用。Among them, S represents the set of all access networks, N represents the total number of access networks, and_Πi represents the user access utility of access network i.

上述方法中，步骤3)包括：In the above-mentioned method, step 3) comprises:

如果所述公平性指数小于预定门限值，则判断所述异构接入网络之间负载失衡，在所述异构接入网络之间执行负载平衡操作。If the fairness index is smaller than the predetermined threshold value, it is judged that the load among the heterogeneous access networks is unbalanced, and a load balancing operation is performed among the heterogeneous access networks.

上述方法中，在步骤3)中，在所述异构接入网络之间执行负载平衡操作包括：In the above method, in step 3), performing a load balancing operation between the heterogeneous access networks includes:

步骤A)、计算各接入网络达到负载均衡的用户数。Step A), calculating the number of users whose load is balanced in each access network.

步骤B)、在所有异构接入网络中，对于接入用户数小于该接入网络达到负载均衡的用户数并且该接入网络的当前网络偏移值小于预定阈值的接入网络，增加该接入网络的当前网络偏移值并广播给用户，直到该接入网络的接入用户数大于等于其达到负载均衡的用户数或者该接入网络的当前网络偏移值大于等于预定阈值；其中，用户根据接入网络的当前网络偏移值和参考信号接收功率选择接入网络。Step B), in all heterogeneous access networks, for access networks whose number of access users is less than the number of users achieving load balancing of the access network and whose current network offset value of the access network is less than a predetermined threshold, increase the The current network offset value of the access network is broadcast to the users until the number of access users of the access network is greater than or equal to the number of load-balanced users or the current network offset value of the access network is greater than or equal to a predetermined threshold; where , the user selects an access network according to the current network offset value of the access network and the received power of the reference signal.

上述方法中，在步骤A)中，计算各接入网络达到负载均衡的用户数包括：In the above method, in step A), calculating the number of users that each access network reaches load balancing includes:

步骤i)、对于各接入网络求解以下复制动态方程组，得到该接入网络中用户占所有用户的比例的稳定解：Step i), solving the following replicated dynamic equations for each access network to obtain a stable solution of the proportion of users in the access network to all users:

$\frac{\mathrm{<span><span>d</span>d</span>}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}}{\mathrm{<span><span>dt</span>dt</span>}}<span><span>=</span>=</span>\mathrm{<span><span>\&delta;</span>\&delta;</span>}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>\&CenterDot;</span>\&CenterDot;</span><span><span>(</span>(</span>{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>-</span>-</span>\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>)</span>)</span><span><span>=</span>=</span>\mathrm{<span><span>0</span>0</span>}<span><span>,</span>,</span><span><span>\&ForAll;</span>\&ForAll;</span>\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}$

$\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}$

其中，S表示所有接入网络的集合，x_i(t)表示在第t次迭代接入网络i中用户占所有用户的比例，∏_i(t)表示在第t次迭代接入网络i的用户接入效用，表示在第t次迭代所有接入网络的用户平均接入效用，t表示迭代次数；Among them, S represents the set of all access networks, xi₍ t) represents the proportion of users accessing network i in the t-th iteration, ∏_i (t) represents the number of users accessing the network i in the t-th iteration user access utility, Represents the average access utility of all users accessing the network at the t-th iteration, and t represents the number of iterations;

步骤ii)、对于各接入网络，将该接入网络中用户占所有用户的比例的稳定解乘以所有用户的数量得到该接入网络达到负载均衡的用户数。Step ii), for each access network, multiply the stable solution of the ratio of users in the access network to all users by the number of all users to obtain the number of users in the access network that achieves load balance.

上述方法中，根据下式计算所有接入网络的用户平均接入效用：In the above method, the average access utility of all users accessing the network is calculated according to the following formula:

$\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}<span><span>=</span>=</span>\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}$

其中，S表示所有接入网络的集合，Π_i表示接入网络i的用户接入效用，x_i表示接入网络i中用户占所有用户的比例。Among them, S represents the set of all access networks,_Πi represents the user access utility of access network i, and_xi represents the proportion of users in access network i to all users.

上述方法中，步骤B)包括对于各接入网络执行如下操作：In the above method, step B) includes performing the following operations for each access network:

步骤B1)、该接入网络统计接入用户数；Step B1), the access network counts the number of access users;

步骤B2)、如果接入用户数小于该接入网络达到负载均衡的用户数并且当前网络偏移值小于预定阈值，则该接入网络将当前网络偏移值增加预定值并向用户广播当前网络偏移值，返回步骤B1)；其中，用户根据接入网络的当前网络偏移值和参考信号接收功率选择接入网络。Step B2), if the number of access users is less than the number of users that the access network achieves load balancing and the current network offset value is less than a predetermined threshold, the access network increases the current network offset value by a predetermined value and broadcasts the current network offset value to users. offset value, return to step B1); wherein, the user selects an access network according to the current network offset value of the access network and the received power of the reference signal.

在一个实施例中，所述方法还包括：In one embodiment, the method also includes:

对于各接入网络，将该接入网络的当前网络偏移值减去所有异构接入网络的当前网络偏移值中的最小值作为该接入网络的当前网络偏移值。For each access network, the current network offset value of the access network minus the minimum value of the current network offset values of all heterogeneous access networks is used as the current network offset value of the access network.

在一个实施例中，所述方法还包括：In one embodiment, the method also includes:

步骤4)、在经历预定时段后，返回步骤1)。Step 4), after a predetermined period of time, return to step 1).

上述方法中，在步骤B)中，用户根据接入网络的当前网络偏移值和参考信号接收功率选择接入网络包括：In the above method, in step B), the user selects the access network according to the current network offset value and the reference signal received power of the access network, including:

用户选择该用户允许接入网络中当前网络偏移值与参考信号接收功率之和最大的接入网络。The user selects the access network with the largest sum of the current network offset value and the reference signal received power among the access networks allowed by the user.

上述方法中，在步骤1)之前还包括：In the above method, before step 1), it also includes:

初始化各接入网络的当前网络偏移值；Initialize the current network offset value of each access network;

各接入网络向用户广播当前网络偏移值；其中，用户根据接入网络的当前网络偏移值和参考信号接收功率选择接入网络。Each access network broadcasts the current network offset value to the user; wherein, the user selects the access network according to the current network offset value of the access network and the received power of the reference signal.

本发明具有如下有益效果：The present invention has following beneficial effect:

通过定义合适的指标来衡量网络负载的失衡情况，只有当失衡超出预先给定的门限时，才会进行负载均衡。此外，设定用户终端接入网络选择方法为选择用户终端允许接入无线网络中参考信号接收功率加接入网络偏移值最大的网络，并且基于接入网络与中心控制单元的信息交互动态调节接入网络偏移值以达到异构接入网络间负载均衡的目的。将区域内所有接入网络具有相同的用户接入效用看作理想的区域达到负载均衡的状态。中心控制单元根据网络状况一次计算得到负载均衡状态的各接入网络的用户接入数，基于上述用户接入选择机制，通过增大低负载网络的网络偏移值，吸引其周围用户接入低负载区域以达到高负载网络用户向低负载网络的转移。本发明减少了负载均衡过程中接入网络与中心控制器以及用户与接入网络的信令交互次数，降低了信令负荷，并且简化了用户接入网络的选择过程。By defining appropriate indicators to measure the imbalance of network load, only when the imbalance exceeds the predetermined threshold, load balancing will be performed. In addition, the user terminal access network selection method is set to select the network with the largest reference signal received power plus access network offset value in the wireless network that the user terminal is allowed to access, and dynamically adjust based on the information interaction between the access network and the central control unit Access network offset value to achieve load balancing among heterogeneous access networks. Considering that all access networks in the area have the same user access utility, it is considered that the ideal area achieves a load-balanced state. The central control unit calculates the number of users accessing each access network in a load-balanced state based on the network conditions at one time. Based on the above-mentioned user access selection mechanism, by increasing the network offset value of the low-load network, it attracts surrounding users to access the low-level network. load area to achieve the transfer of high-load network users to low-load network. The invention reduces the number of signaling interactions between the access network and the central controller and between the user and the access network during the load balancing process, reduces the signaling load, and simplifies the selection process of the user access network.

附图说明Description of drawings

以下附图仅对本发明做示意性说明和解释，并不用于限定本发明的范围，其中：The following drawings only schematically illustrate and explain the present invention, and are not intended to limit the scope of the present invention, wherein:

图1是根据本发明一个实施例的异构接入网络之间负载均衡方法的示意图；FIG. 1 is a schematic diagram of a load balancing method between heterogeneous access networks according to an embodiment of the present invention;

图2是现有的一种异构接入网络之间负载均衡方法的示意图。Fig. 2 is a schematic diagram of an existing load balancing method among heterogeneous access networks.

具体实施方式Detailed ways

为了对本发明的技术特征、目的和效果有更加清楚的理解，现对照附图说明本发明的具体实施方式。In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the specific implementation manners of the present invention will now be described with reference to the accompanying drawings.

根据本发明的一个实施例，提供一种异构接入网络之间负载均衡的方法。概括而言，该方法包括：计算所述异构接入网络中各接入网络的用户接入效用，其中用户接入效用用于衡量接入网络的负载；根据所述用户接入效用计算公平性指数；以及，根据所述公平性指数判断是否进行负载平衡。According to an embodiment of the present invention, a load balancing method among heterogeneous access networks is provided. Generally speaking, the method includes: calculating the user access utility of each access network in the heterogeneous access network, wherein the user access utility is used to measure the load of the access network; calculating the fairness based on the user access utility and determining whether to perform load balancing according to the fairness index.

现结合图1，对本发明提供的异构接入网络之间负载均衡的方法进行详细描述。Now, with reference to FIG. 1 , the method for load balancing between heterogeneous access networks provided by the present invention will be described in detail.

步骤S101)、初始化接入网络偏移值。Step S101), initialize the access network offset value.

在初始化时，中心控制单元将区域内所有接入网络的偏移值(例如，接入网络i的偏移值可以用bias_i表示)设置为0，并将各接入网络的偏移值通过回传链路发送给不同的异构接入网络。作为本领域的公知常识，本文中出现的中心控制单元为具有控制功能的抽象服务单元，其可以集中在服务网关或其它核心网络设备中。During initialization, the central control unit sets the bias values of all access networks in the area (for example, the bias value of access network i can be represented by bias_i ) to 0, and the bias values of each access network are passed through The backhaul link is sent to different heterogeneous access networks. As common knowledge in the field, the central control unit presented herein is an abstract service unit with control functions, which can be concentrated in a service gateway or other core network devices.

初始的接入网络偏移值由接入网络通过系统消息广播发送给用户。在一个实施例中，用户选择该用户允许接入网络中参考信号接收功率加接入网络偏移值最大的网络作为接入网络。由于初始时各接入网络的偏移值为0，则这时的用户选择过程相当于用户按照传统的选择最大参考信号接收功率的方式选择接入网络。The initial access network offset value is sent to the user by the access network through system message broadcast. In an embodiment, the user selects the network with the largest reference signal received power plus access network offset value among the access networks allowed by the user as the access network. Since the offset value of each access network is 0 initially, the user selection process at this time is equivalent to the user selecting an access network in a traditional way of selecting the maximum reference signal received power.

本领域技术人员应理解，参考信号接收功率可通过用户终端的信号测量过程获得。测量是指基站或接入点向终端发送测量请求，终端根据测量要求进行特定网络的信号测量，即获得用户终端接入所要求测量小区的参考信号接收功率。在一个实施例中，基站在检测到用户接入服务小区的信号强度高于/低于门限或者相邻小区信号强度高于门限，以及服务基站信号强度低于门限而相邻基站信号高于门限等情况下均可以发起测量请求。在用户初始接入网络时，接入网络的接入点(基站或其它无线接入点)会和用户终端通过信令交互，实现终端和接入网络的连接，这个过程中就包含了测量；另外，在用户进行切换之前，基站也会请求用户终端进行测量。Those skilled in the art should understand that the received power of the reference signal can be obtained through a signal measurement process of the user terminal. Measurement means that the base station or access point sends a measurement request to the terminal, and the terminal performs signal measurement of a specific network according to the measurement request, that is, obtains the reference signal received power of the measurement cell required for the user terminal to access. In one embodiment, the base station detects that the signal strength of the user accessing the serving cell is higher/lower than the threshold or the signal strength of the neighboring cell is higher than the threshold, and the signal strength of the serving base station is lower than the threshold and the signal strength of the neighboring base station is higher than the threshold In other cases, a measurement request can be initiated. When the user initially accesses the network, the access point (base station or other wireless access point) of the access network will interact with the user terminal through signaling to realize the connection between the terminal and the access network. This process includes measurement; In addition, before the user performs handover, the base station will also request the user terminal to perform measurement.

步骤S102)、启动负载均衡定时器。Step S102), start a load balancing timer.

中心控制单元启动负载均衡定时器，开始负载均衡周期计时。The central control unit starts the load balancing timer, and starts the timing of the load balancing cycle.

步骤S103)、通过判断负载均衡定时器是否达到预先设定的负载均衡周期时长，确定是否进行下一步。Step S103), determine whether to proceed to the next step by judging whether the load balancing timer reaches the preset load balancing cycle duration.

中心控制单元判断负载均衡定时器是否达到预先设定的负载均衡周期时长；若达到预先设定的负载均衡周期时长，则中心控制单元停止负载均衡定时器，进入下一步来判断是否需要进行负载均衡；否则，继续执行步骤S103)。The central control unit judges whether the load balancing timer reaches the preset load balancing cycle duration; if it reaches the preset load balancing cycle duration, the central control unit stops the load balancing timer and enters the next step to judge whether load balancing is required ; Otherwise, continue to execute step S103).

步骤S104)、衡量网络的负载。Step S104), measuring the load of the network.

在一个实施例中，接入网络根据用户接入网络的效用值(也称作用户接入效用值，现有技术中通常以接入用户的平均吞吐量作为用户接入效用值)来衡量该接入网络的负载。其中，接入网络中的用户接入效用值越高，该接入网络负载越低。In one embodiment, the access network measures the network according to the utility value of the user's access to the network (also referred to as the user access utility value, in the prior art, the average throughput of the access user is usually used as the user access utility value). Access network load. Wherein, the higher the user access utility value in the access network, the lower the load of the access network.

在一个优选的实施例中，定义用户接入网络的效用值为该网络中用户的平均吞吐量，即网络总带宽容量与该网络中接入用户数的比值。例如，在异构接入网络服务区域A内，接入网络i可首先统计接入用户数，并且根据下式计算用户接入效用：In a preferred embodiment, the utility value of user access to the network is defined as the average throughput of users in the network, that is, the ratio of the total bandwidth capacity of the network to the number of access users in the network. For example, in the heterogeneous access network service area A, the access network i can first count the number of access users, and calculate the user access utility according to the following formula:

∏_i＝R_i(x_i)          (1)∏_i ＝R_i (_xi ) (1)

其中，Π_i表示接入网络i的用户接入效用；x_i表示接入网络i中的用户占区域A内所有用户的比例，x_i＝n_i/K，其中n_i表示接入网络i中接入用户数，K表示区域A中所有网络接入用户的总数，S是区域A中所有的异构接入网络集合；R_i(x_i)＝C_i/n_i＝C_i/(x_i·K)表示接入网络i中用户得到的平均吞吐量，C_i表示接入网络i的总带宽容量，其单位为bps。Among them, Π_i represents the user access utility of access network i;_xi represents the proportion of users in access network i to all users in area A,_xi = n_i /K, where n_i represents access network i The number of users accessing the network, K represents the total number of users accessing the network in area A, S is the set of all heterogeneous access networks in area A; R_i (xi₎ =C_i /n_i =C_i /(_xi ·K) represents the average throughput obtained by users in access network i, C_i represents the total bandwidth capacity of the access network i, and its unit is bps.

在其他实施例中，也可以将用户接入基站的信号质量，或者基站中用户的中断概率等作为接入网络中的用户接入效用。In other embodiments, the signal quality of the user accessing the base station, or the outage probability of the user in the base station, etc. may also be used as the user access utility in the access network.

步骤S105)、每个接入网络将该接入网络的用户接入效用值和接入用户数通过回传链路发送给中心控制单元。Step S105), each access network sends the user access utility value and the number of access users of the access network to the central control unit through the backhaul link.

步骤S106)、中心控制单元判断是否进行负载均衡。Step S106), the central control unit judges whether to perform load balancing.

中心控制单元计算接入网络之间的负载公平性指数，用于衡量网络负载失衡的程度。在一个实施例中，采用不同接入网络之间用户接入效用的Jain’s公平性指数作为区域内接入网络间的负载公平性指数(后文或简称为公平性指数)。其中，中心控制单元根据区域A内每个接入网络的用户接入效用值计算不同接入网络之间的公平性指数，如果该公平性指数大于等于预先指定的门限值，则不进行负载均衡并且返回步骤S102)，否则，进入步骤S107)。The central control unit calculates the load fairness index between access networks, which is used to measure the degree of network load imbalance. In one embodiment, the Jain's fairness index of user access utility between different access networks is used as the load fairness index among the access networks in the area (hereinafter or simply referred to as the fairness index). Among them, the central control unit calculates the fairness index between different access networks according to the user access utility value of each access network in area A, and if the fairness index is greater than or equal to a pre-specified threshold value, no load Balance and return to step S102), otherwise, enter step S107).

根据本发明的一个实施例，中心控制单元可根据如下公式来计算区域A内接入网络之间的公平性指数LFI：According to an embodiment of the present invention, the central control unit can calculate the fairness index LFI between access networks in area A according to the following formula:

$\mathrm{<span><span>LFI</span>LFI</span>}<span><span>=</span>=</span>{<span><span>(</span>(</span>\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>)</span>)</span>}^{\mathrm{<span><span>2</span>2</span>}}<span><span>/</span>/</span><span><span>(</span>(</span>\mathrm{<span><span>N</span>N</span>}\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}<span><span>(</span>(</span>{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}^{\mathrm{<span><span>2</span>2</span>}}<span><span>)</span>)</span><span><span>)</span>)</span><span><span>-</span>-</span><span><span>-</span>-</span><span><span>-</span>-</span><span><span>(</span>(</span>\mathrm{<span><span>2</span>2</span>}<span><span>)</span>)</span>$

其中，如上文所述，S为区域A内所有异构接入网络的集合，N为区域A内所有接入网络的总数，Π_i为接入网络i的用户接入效用。Among them, as mentioned above, S is the set of all heterogeneous access networks in area A, N is the total number of all access networks in area A, and_Πi is the user access utility of access network i.

在其他实施例中，也可以采用以下方式来判断是否需要进行区域A的负载均衡。例如，计算区域A内各接入网络中接入用户最多的网络接入用户数与接入用户最少的网络接入用户数的差值，来作为衡量网络间负载失衡的依据；将各接入网络接入用户数与网络平均接入用户数之差的绝对值中的最大值，作为衡量网络间负载失衡的依据；或者，采用区域A内所有接入网络中中断用户(即发送速率吞吐量低于最小需要速率的用户)的总数来衡量网络间负载失衡。In other embodiments, the following manner may also be used to determine whether to perform load balancing in area A. For example, calculate the difference between the number of network access users with the largest number of access users and the number of network access users with the least access users in each access network in area A, as a basis for measuring the load imbalance between networks; The maximum value of the absolute value of the difference between the number of network access users and the average number of network access users is used as the basis for measuring the load imbalance between networks; or, the interrupt users in all access networks in area A (that is, the transmission rate throughput The total number of users below the minimum required rate) to measure the load imbalance between networks.

在本步骤中设置一定的公平性指数门限值，只有接入网络之间的负载公平性指数低于该门限值才进行负载均衡，这样做降低了信令负荷。In this step, a certain fairness index threshold value is set, and load balancing is performed only when the load fairness index between access networks is lower than the threshold value, which reduces signaling load.

步骤S107)、中心控制单元计算区域内所有接入网络中用户平均接入效用值。Step S107), the central control unit calculates the average access utility value of users in all access networks in the area.

在一个实施例中，中心控制单元根据如下公式计算区域A内所有接入网络中用户平均接入效用In one embodiment, the central control unit calculates the average access utility of users in all access networks in area A according to the following formula

$\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}<span><span>=</span>=</span>\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>-</span>-</span><span><span>-</span>-</span><span><span>-</span>-</span><span><span>(</span>(</span>\mathrm{<span><span>3</span>3</span>}<span><span>)</span>)</span>$

其中，S为区域A内所有异构接入网络的集合，x_i为接入网络i中的用户占区域A内所有用户的比例，Π_i为接入网络i的用户接入效用。Among them, S is the set of all heterogeneous access networks in area A,_xi is the proportion of users in access network i to all users in area A, and_Πi is the user access utility of access network i.

步骤S108)、将区域内所有接入网络具有相同的用户接入效用看作理想的区域达到负载均衡的状态，由中心控制单元计算区域达到负载均衡状态的各接入网络的用户数。Step S108), consider that all access networks in the area have the same user access utility as an ideal area reaching a load-balanced state, and the central control unit calculates the number of users of each access network that has reached a load-balanced state in the area.

在一个优选的实施例中，中心控制单元可根据区域内不同接入网络中的用户接入效用Π_i以及接入用户数求解以下复制动态方程组：In a preferred embodiment, the central control unit can solve the following replication dynamic equations according to the user access utility Π_i and the number of access users in different access networks in the area:

$\frac{\mathrm{<span><span>d</span>d</span>}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}}{\mathrm{<span><span>dt</span>dt</span>}}<span><span>=</span>=</span>\mathrm{<span><span>\&delta;</span>\&delta;</span>}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>\&CenterDot;</span>\&CenterDot;</span><span><span>(</span>(</span>{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>-</span>-</span>\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>)</span>)</span><span><span>=</span>=</span>\mathrm{<span><span>0</span>0</span>}<span><span>,</span>,</span><span><span>\&ForAll;</span>\&ForAll;</span>\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}$

$\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}<span><span>-</span>-</span><span><span>-</span>-</span><span><span>-</span>-</span><span><span>(</span>(</span>\mathrm{<span><span>4.2</span>4.2</span>}<span><span>)</span>)</span>$

其中，S表示所有接入网络的集合，x_i(t)表示在第t次迭代接入网络i中用户占所有用户的比例，∏_i(t)表示在第t次迭代接入网络i的用户接入效用，表示在第t次迭代所有接入网络的用户平均接入效用，t表示迭代次数。方程式(4.1)表示随着迭代次数的增长接入网络i的用户比例不再变化，方程式(4.2)表示区域A中所有接入网络的用户占的比例总和为1。Among them, S represents the set of all access networks, xi₍ t) represents the proportion of users accessing network i in the t-th iteration, ∏_i (t) represents the number of users accessing the network i in the t-th iteration user access utility, Represents the average access utility of all users accessing the network at the t-th iteration, and t represents the number of iterations. Equation (4.1) indicates that the proportion of users accessing network i does not change as the number of iterations increases, and Equation (4.2) indicates that the sum of the proportions of all users accessing network i in area A is 1.

根据李亚普诺夫(A.M.Lyapunov)系统稳定性分析理论，由于上述复制动态方程的雅可比矩阵的特征值有严格负实部，所以上述复制动态方程组存在稳定解x_i。根据每个接入网络i的达到稳定均衡状态的解x_i，即可求得各接入网络达到负载均衡状态的用户数即K为区域A中所有网络接入用户的总数。According to AMLyapunov's system stability analysis theory, since the eigenvalues of the Jacobian matrix of the above-mentioned replicated dynamic equations have strictly negative real parts, there is a stable solution x_i for the above-mentioned replicated dynamic equations. According to the solution x_i of each access network i reaching a stable and balanced state, the number of users in each access network that has reached a load-balanced state can be obtained Right now K is the total number of all network access users in area A.

在其他实施例中，也可以将区域内所有用户数除以区域内总的网络数得到负载均衡状态的用户数，但这种方法只适用于接入网络带宽相同的情况。In other embodiments, the number of users in the load-balanced state can also be obtained by dividing the number of all users in the area by the total number of networks in the area, but this method is only applicable to the case of the same access network bandwidth.

步骤S109)、中心控制单元将区域内各接入网络达到负载均衡状态的用户数，通过回传链路发送给每个接入网络。Step S109), the central control unit sends the number of users of each access network in the area that has reached the load balance state to each access network through the backhaul link.

步骤S110)、在所有异构接入网络中，对于接入用户数小于该接入网络达到负载均衡的用户数并且该接入网络的网络偏移值小于最大偏移值的接入网络，增加该接入网络的网络偏移值并广播给用户，直到该接入网络的接入用户数大于等于其达到负载均衡的用户数或者网络偏移值大于等于最大偏移值。Step S110), in all heterogeneous access networks, for access networks whose number of access users is less than the number of users achieving load balancing of the access network and whose network offset value is less than the maximum offset value, add The network offset value of the access network is broadcast to users until the number of access users of the access network is greater than or equal to the number of users achieving load balancing or the network offset value is greater than or equal to the maximum offset value.

详细来说，在一个实施例中，区域A中各接入网络i统计其接入用户数n_i，并比较接入用户数n_i和接入网络i达到负载均衡状态的用户数n_i^e。如果ni小于n_i^e则接入网络i增加网络偏移值，即bias_i＝bias_i+delta，并广播该修改的网络偏移值，之后接入网络i再次统计接入用户数，如果ni小于n_i^e且当前bias_i小于最大偏移值的接入网络，则再次增加网络偏移值，直到接入用户数n_i大于等于接入网络i达到负载均衡状态的用户数n_i^e或者接入网络i的网络偏移值bias_i达到最大偏移值(或称最大网络偏移值)，则接入网络i结束本次负载均衡网络偏移调节，进行步骤S111)；否则，如果n_i大于等于n_i^e则直接进行下一步。In detail, in one embodiment, each access network i in area A counts the number of access users n_i , and compares the number of access users n_i with the number of users n_i^e of access network i reaching the load balancing state . If ni is smaller than n_i^e , access network i increases the network offset value, that is, bias_i = bias_i + delta, and broadcasts the modified network offset value, and then access network i counts the number of access users again, if ni If the access network is less than n_i^e and the current bias_i is less than the maximum offset value, then increase the network offset value again until the number of access users n_i is greater than or equal to the number of users accessing network i to achieve load balancing state n_i^e or The network bias value bias i of the access network_i reaches the maximum bias value (or the maximum network bias value), then the access network i ends this load balancing network bias adjustment, and proceeds to step S111); otherwise, if n If_i is greater than or equal to n_i^e , proceed directly to the next step.

在本步骤中，如果增大了接入网络的偏移值，如上文所述，由于用户选择所有可接入网络中参考信号接收功率加网络偏移值最大的网络作为接入网络，则该接入网络内的接入用户数可能会增加，使得用户从高负载区域向低负载区域转移，然而，增大网络偏移会使远离网络中心区域的用户接入，这样会降低用户的信号质量从而降低用户的吞吐量性能。因而，为了在进行接入网络间负载均衡的同时考虑网络内个体用户性能的需要，设置一个最大网络偏移值。当网络偏移达到最大网络偏移值时，不再增大该网络偏移值。在一个实施例中，可根据经验值来设定该最大网络偏移值，如果以db为偏移值的单位，优选地，最大网络偏移值是约4-20db内的一个值。In this step, if the offset value of the access network is increased, as mentioned above, since the user selects the network with the largest reference signal received power plus network offset value among all accessible networks as the access network, the The number of access users in the access network may increase, causing users to transfer from high-load areas to low-load areas. However, increasing the network offset will allow users far from the center of the network to access, which will reduce the signal quality of users Thereby degrading the throughput performance of the user. Therefore, a maximum network offset value is set in order to consider the performance requirements of individual users in the network while performing load balancing among access networks. When the network offset reaches the maximum network offset value, the network offset value is no longer increased. In one embodiment, the maximum network offset value can be set according to empirical values. If db is used as the offset value unit, preferably, the maximum network offset value is a value within about 4-20db.

本步骤通过增大所需负载均衡区域中低负载网络的网络偏移值，吸引其周围用户接入低负载区域以达到高负载网络用户向低负载网络的转移，直到负载均衡区域中各网络接入用户数达到负载均衡状态的用户数，或者网络偏移达到最大不能再进一步吸引相邻网络用户接入。本步骤可以通过较少的信令交互来实现接入网络间的负载均衡。In this step, by increasing the network offset value of the low-load network in the required load-balancing area, it attracts users around it to access the low-load area so as to achieve the transfer of users from the high-load network to the low-load network, until each network in the load-balancing area connects to the low-load area. If the number of incoming users reaches the load-balanced state, or the network offset reaches the maximum, it cannot further attract adjacent network users to access. In this step, load balancing among access networks can be realized through less signaling interaction.

步骤S111)、接入网络将当前网络偏移值通过回传链路发送给中心控制单元。Step S111), the access network sends the current network offset value to the central control unit through the backhaul link.

步骤S112)、中心控制单元在收到区域A内所有接入网络的当前网络偏移值后，计算各接入网络的相对网络偏移值。Step S112), after receiving the current network offset values of all the access networks in the area A, the central control unit calculates the relative network offset values of each access network.

在一个实施例中，接入网络的相对网络偏移值为将该接入网络的当前网络偏移值减去所有接入网络当前偏移值中的最小值。In one embodiment, the relative network offset value of the access network is the minimum value of the current network offset value of the access network minus the current offset values of all access networks.

步骤S113)、中心控制单元将各接入网络的相对网络偏移值通过回传链路发送给各接入网络，将接入网络的相对网络偏移值作为该接入网络当前的网络偏移值，结束本次负载均衡过程，返回步骤S102)。Step S113), the central control unit sends the relative network offset value of each access network to each access network through the backhaul link, and uses the relative network offset value of the access network as the current network offset value of the access network value, end the load balancing process, and return to step S102).

采用相对网络偏移的目的是避免网络偏移值增大到最大后无法进行后续的负载均衡。The purpose of using the relative network offset is to avoid subsequent load balancing after the network offset value increases to the maximum.

以下通过系统仿真实验对本发明提供的异构接入网络之间负载均衡的方法与传统用户接入方法的性能进行对比。The performance of the load balancing method between heterogeneous access networks provided by the present invention and the traditional user access method is compared through system simulation experiments below.

该仿真实验的区域是一个LTE网络区域中覆盖1个WLAN网络的异构网络区域A，不同接入网络的相关参数配置参见表1，并且定义了5种用户分布场景，参见表2。此外，图2描述了传统的异构接入网络负载均衡方法的示意图。The area of the simulation experiment is a heterogeneous network area A covering one WLAN network in an LTE network area. The relevant parameter configurations of different access networks are shown in Table 1, and 5 user distribution scenarios are defined, see Table 2. In addition, FIG. 2 depicts a schematic diagram of a traditional load balancing method for a heterogeneous access network.

表1Table 1

参数parameter值value系统带宽system bandwidthLTE和WLAN都为5MHZBoth LTE and WLAN are 5MHZLTE网络数Number of LTE networks11WLAN网络区域WLAN network area11LTE区域半径LTE area radius500m500m

WLAN区域半径WLAN area radius35m35mMacrocell天线增益Macrocell Antenna Gain14dbi14dbiWLAN天线增益WLAN antenna gain5dbi5dbi基站天线数Number of base station antennas1发送1接收1 send 1 receiveMacrocell到UE路损Macrocell to UE path loss128.1+37.6*log10(R)R单位为km128.1+37.6*log10(R)R unit is kmWLAN到UE的路损Path loss from WLAN to UE140.7+36.7*log10(R)R单位为km140.7+36.7*log10(R)R unit is km阴影衰落shadow fading对数正态阴影衰落的标准差8dbStandard deviation 8db of lognormal shadow fading

表2Table 2

场景ScenesLTE区域用户数Number of users in LTE areaWLAN区域用户数Number of users in the WLAN area场景1scene 110101010场景2scene 210102020场景3scene 310103030场景4scene 410104040场景5scene 510105050

表3给出了在场景5中本发明提供的方法与传统方法(或称D.Niyato的方法)的操作对比。通过分析可以看出本发明提供的方法简化了用户接入网络的选择过程、简化了用户与接入网络的交互、减少了接入网络的计算操作并减少了接入网络与中心控制器的交互次数，并且减少了中心控制器的计算操作。Table 3 shows the operation comparison between the method provided by the present invention and the traditional method (or D. Niyato's method) in Scenario 5. Through the analysis, it can be seen that the method provided by the present invention simplifies the selection process of the user access network, simplifies the interaction between the user and the access network, reduces the calculation operation of the access network, and reduces the interaction between the access network and the central controller times, and reduce the computing operations of the central controller.

表3table 3

因此，本发明提供的方法减少了负载均衡过程的信令交互次数。此外，简化了用户接入网络的选择过程，在降低用户接入时延的同时也简化了异构网络之间负载均衡的过程。Therefore, the method provided by the present invention reduces the number of signaling interactions in the load balancing process. In addition, the selection process of the user access network is simplified, and the process of load balancing between heterogeneous networks is also simplified while reducing the user access delay.

下文对比了在四种用户接入方法下，用户的几何平均吞吐量曲线以及用户的公平性指数。其中，四种方法分别是：The following compares the user's geometric mean throughput curve and the user's fairness index under the four user access methods. Among them, the four methods are:

1、选择最大信号强度网络接入的方法；1. Select the method of network access with maximum signal strength;

2、选择最近距离网络接入的方法；2. Select the nearest network access method;

3、D.Niyato提出的负载均衡方法；和3. The load balancing method proposed by D.Niyato; and

4、本发明提供的异构接入网络之间负载均衡的方法。4. The load balancing method among heterogeneous access networks provided by the present invention.

其中，按照下式计算用户的几何平均吞吐量：Among them, the user's geometric mean throughput is calculated according to the following formula:

$\sqrt[<span><span>|</span>|</span>\mathrm{<span><span>U</span>u</span>}<span><span>|</span>|</span>]{{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>u</span>u</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>U</span>u</span>}}{\mathrm{<span><span>r</span>r</span>}}_{\mathrm{<span><span>u</span>u</span>}}}<span><span>-</span>-</span><span><span>-</span>-</span><span><span>-</span>-</span><span><span>(</span>(</span>\mathrm{<span><span>5</span>5</span>}<span><span>)</span>)</span>$

其中r_u表示用户u的平均吞吐量，U表示区域A内所有用户集，|U|表示区域A内所有用户的总数。where r_u represents the average throughput of user u, U represents the set of all users in area A, and |U| represents the total number of all users in area A.

另外，用户吞吐量公平性指数表示为：In addition, the user throughput fairness index is expressed as:

$\frac{{<span><span>(</span>(</span>{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}_{\mathrm{<span><span>u</span>u</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>U</span>u</span>}}{\mathrm{<span><span>r</span>r</span>}}_{\mathrm{<span><span>u</span>u</span>}}<span><span>)</span>)</span>}^{\mathrm{<span><span>2</span>2</span>}}}{<span><span>|</span>|</span>\mathrm{<span><span>U</span>u</span>}<span><span>|</span>|</span>{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}_{\mathrm{<span><span>u</span>u</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>U</span>u</span>}}{{\mathrm{<span><span>r</span>r</span>}}_{\mathrm{<span><span>u</span>u</span>}}}^{\mathrm{<span><span>2</span>2</span>}}}<span><span>-</span>-</span><span><span>-</span>-</span><span><span>-</span>-</span><span><span>(</span>(</span>\mathrm{<span><span>6</span>6</span>}<span><span>)</span>)</span>$

该式反映了接入网络对各服务用户的公平性。This formula reflects the fairness of the access network to each service user.

表4给出了在不同场景下采用不同接入方法的用户几何平均吞吐量。从表4可以看出，在五种不同的用户分布场景下，采用本发明提供的异构接入网络之间负载均衡的方法和D.Niyato的方法，用户几何平均吞吐量明显高于其他两种方法，其中选择最近距离网络接入方法的用户几何平均吞吐量最差。这是因为，用户选择最近距离接入会导致大量的用户接入WLAN热点区域，造成局部拥挤；而采用最大信号强度网络接入方法，由于LTE基站发射功率高于WLAN基站，WLAN热点区域边缘的用户会选择接入LTE基站，在一定程度上平衡了两个接入网络的负载。本发明提供的方法略优于D.Niyato的方法，这是因为本发明的方法通过逐渐增加高效用值网络的网络偏移，使距离高效用值网络较近的用户接入，从而进行高负载网络用户向低负载网络的转移，这样做优于D.Niyato的方法中低效用值网络内用户随机选择接入高效用值网络。Table 4 shows the geometric mean throughput of users using different access methods in different scenarios. It can be seen from Table 4 that under five different user distribution scenarios, the geometric average throughput of users is significantly higher than that of the other two by adopting the method of load balancing between heterogeneous access networks provided by the present invention and the method of D.Niyato. A method, in which the geometric average throughput of users who choose the shortest distance network access method is the worst. This is because, if the user selects the shortest distance to access, a large number of users will access the WLAN hotspot area, causing local congestion; while using the maximum signal strength network access method, since the transmit power of the LTE base station is higher than that of the WLAN base station, the WLAN hotspot area edge Users will choose to access LTE base stations, which balances the load of the two access networks to a certain extent. The method provided by the present invention is slightly better than the method of D.Niyato, because the method of the present invention gradually increases the network offset of the high-utility value network, so that users who are closer to the high-utility value network can access, thereby performing high-load The transfer of network users to low-load networks is better than D. Niyato's method in which users in low-utility networks randomly choose to access high-utility networks.

表4Table 4

表5给出了在不同场景下采用不同接入方法的用户间公平性指数。从表5可以看出在五种不同的用户分布场景下，采用本发明提供的异构接入网路之间负载均衡的方法和D.Niyato的方法，用户吞吐量公平性指数明显高于其他两种方法，其中最近距离网络接入方法的用户间公平性指数最小。这是因为用户选择最近距离接入会导致大量的用户接入WLAN热点区域，造成局部拥挤；而采用最大信号强度网络接入方法，由于LTE基站发射功率高于WLAN基站，WLAN热点区域边缘的用户会选择接入LTE基站，一定程度上平衡了两个接入网络的负载。由于本发明提供的方法与D.Niyato的方法都进行了网络间负载均衡的用户接入处理，因此采用本发明提供的方法得到的用户间吞吐量公平性指数与D.Niyato的方法接近。Table 5 shows the fairness index among users using different access methods in different scenarios. It can be seen from Table 5 that under five different user distribution scenarios, the user throughput fairness index is obviously higher than other methods by adopting the method of load balancing between heterogeneous access networks provided by the present invention and the method of D.Niyato Among the two methods, the fairness index among users of the closest distance network access method is the smallest. This is because the user chooses the shortest distance access will cause a large number of users to access the WLAN hotspot area, resulting in local congestion; while using the maximum signal strength network access method, because the transmit power of the LTE base station is higher than that of the WLAN base station, the users at the edge of the WLAN hotspot area It will choose to access the LTE base station, which balances the load of the two access networks to a certain extent. Since both the method provided by the present invention and the method of D. Niyato perform user access processing of inter-network load balancing, the throughput fairness index among users obtained by the method provided by the present invention is close to that of D. Niyato's method.

表5table 5

不同接入方法different access methods场景1scene 1场景2scene 2场景3scene 3场景4scene 4场景5scene 5最大信号强度接入Maximum Signal Strength Access0.800.800.950.950.900.900.880.880.880.88距离最近接入closest access0.740.740.720.720.500.500.500.500.460.46D.Niyato的方法D. Niyato's method0.920.920.950.950.950.950.970.970.960.96本发明提出的方法The method proposed by the present invention0.920.920.950.950.960.960.970.970.970.97

以上仿真实验的结果说明了采用本发明提供的异构接入网络之间负载均衡的方法与采用D.Niyato的负载均衡方法以及其他传统的用户接入方法相比，能够提高异构接入网络中用户资源分配的公平性，从而达到异构网络间负载均衡的目的。The result of the above simulation experiment has illustrated that adopting the method for load balancing between the heterogeneous access networks provided by the present invention compared with the load balancing method and other traditional user access methods using D.Niyato can improve the performance of the heterogeneous access network. In order to achieve the fairness of user resource allocation in the network, the purpose of load balancing among heterogeneous networks is achieved.

应当理解，虽然本说明书是按照各个实施例描述的，但并非每个实施例仅包含一个独立的技术方案，说明书的这种叙述方式仅仅是为清楚起见，本领域技术人员应当将说明书作为一个整体，各实施例中的技术方案也可以经适当组合，形成本领域技术人员可以理解的其他实施方式。It should be understood that although this description is described according to various embodiments, not each embodiment only includes an independent technical solution, and this description of the description is only for clarity, and those skilled in the art should take the description as a whole , the technical solutions in the various embodiments can also be properly combined to form other implementations that can be understood by those skilled in the art.

以上所述仅为本发明示意性的具体实施方式，并非用以限定本发明的范围。任何本领域的技术人员，在不脱离本发明的构思和原则的前提下所作的等同变化、修改与结合，均应属于本发明保护的范围。The above descriptions are only illustrative specific implementations of the present invention, and are not intended to limit the scope of the present invention. Any equivalent changes, modifications and combinations made by those skilled in the art without departing from the concept and principle of the present invention shall fall within the protection scope of the present invention.

Claims (13)

Translated fromChinese

1.一种异构接入网络之间负载均衡的方法，包括：1. A method for load balancing between heterogeneous access networks, comprising:步骤1)、计算所述异构接入网络中各接入网络的用户接入效用；其中，用户接入效用用于衡量接入网络的负载；Step 1), calculating the user access utility of each access network in the heterogeneous access network; wherein, the user access utility is used to measure the load of the access network;步骤2)、根据所述用户接入效用计算公平性指数；Step 2), calculating a fairness index according to the user access utility;步骤3)、根据所述公平性指数判断是否进行负载平衡。Step 3), judging whether to perform load balancing according to the fairness index.2.根据权利要求1所述的方法，其中，步骤1)包括：2. The method according to claim 1, wherein step 1) comprises:各接入网络统计接入用户数并计算用户接入效用。Each access network counts the number of access users and calculates the user access utility.3.根据权利要求1或2所述的方法，其中，所述用户接入效用是接入网络中用户的平均吞吐量。3. The method according to claim 1 or 2, wherein the user access utility is the average throughput of users in the access network.4.根据权利要求1或2所述的方法，其中，步骤2)包括：4. The method according to claim 1 or 2, wherein, step 2) comprises:计算不同接入网络之间用户接入效用的Jain’s公平性指数，其中，根据下式计算该公平性指数LFI：Calculate the Jain's fairness index of user access utility between different access networks, where the fairness index LFI is calculated according to the following formula:$\mathrm{<span><span>LFI</span>LFI</span>}<span><span>=</span>=</span>{<span><span>(</span>(</span>\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>)</span>)</span>}^{\mathrm{<span><span>2</span>2</span>}}<span><span>/</span>/</span><span><span>(</span>(</span>\mathrm{<span><span>N</span>N</span>}\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}<span><span>(</span>(</span>{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}^{\mathrm{<span><span>2</span>2</span>}}<span><span>)</span>)</span><span><span>)</span>)</span>$其中，S表示所有接入网络的集合，N表示接入网络的总数，Π_i表示接入网络i的用户接入效用。Among them, S represents the set of all access networks, N represents the total number of access networks, and_Πi represents the user access utility of access network i.5.根据权利要求1或2所述的方法，其中，步骤3)包括：5. The method according to claim 1 or 2, wherein, step 3) comprises:如果所述公平性指数小于预定门限值，则判断所述异构接入网络之间负载失衡，在所述异构接入网络之间执行负载平衡操作。If the fairness index is smaller than the predetermined threshold value, it is judged that the load among the heterogeneous access networks is unbalanced, and a load balancing operation is performed among the heterogeneous access networks.6.根据权利要求5所述的方法，在步骤3)中，在所述异构接入网络之间执行负载平衡操作包括：6. The method according to claim 5, in step 3), performing a load balancing operation between the heterogeneous access networks comprises:步骤A)、计算各接入网络达到负载均衡的用户数。Step A), calculating the number of users whose load is balanced in each access network.步骤B)、在所有异构接入网络中，对于接入用户数小于该接入网络达到负载均衡的用户数并且该接入网络的当前网络偏移值小于预定阈值的接入网络，增加该接入网络的当前网络偏移值并广播给用户，直到该接入网络的接入用户数大于等于其达到负载均衡的用户数或者该接入网络的当前网络偏移值大于等于预定阈值；其中，用户根据接入网络的当前网络偏移值和参考信号接收功率选择接入网络。Step B), in all heterogeneous access networks, for access networks whose number of access users is less than the number of users achieving load balancing of the access network and whose current network offset value of the access network is less than a predetermined threshold, increase the The current network offset value of the access network is broadcast to the users until the number of access users of the access network is greater than or equal to the number of load-balanced users or the current network offset value of the access network is greater than or equal to a predetermined threshold; where , the user selects an access network according to the current network offset value of the access network and the received power of the reference signal.7.根据权利要求6所述的方法，其中，在步骤A)中，计算各接入网络达到负载均衡的用户数包括：7. The method according to claim 6, wherein, in step A), calculating the number of users that each access network reaches load balancing comprises:步骤i)、对于各接入网络求解以下复制动态方程组，得到该接入网络中用户占所有用户的比例的稳定解：Step i), solving the following replicated dynamic equations for each access network to obtain a stable solution of the proportion of users in the access network to all users:$\frac{{\mathrm{<span><span>dx</span>dx</span>}}_{\mathrm{<span><span>i</span>i</span>}}}{\mathrm{<span><span>dt</span>dt</span>}}<span><span>=</span>=</span>\mathrm{<span><span>\&delta;</span>\&delta;</span>}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>\&CenterDot;</span>\&CenterDot;</span><span><span>(</span>(</span>{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>-</span>-</span>\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>t</span>t</span>}<span><span>)</span>)</span><span><span>)</span>)</span><span><span>=</span>=</span>\mathrm{<span><span>0</span>0</span>}<span><span>,</span>,</span><span><span>\&ForAll;</span>\&ForAll;</span>\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}$$\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}$其中，S表示所有接入网络的集合，x_i(t)表示在第t次迭代接入网络i中用户占所有用户的比例，Π_i(t)表示在第t次迭代接入网络i的用户接入效用，表示在第t次迭代所有接入网络的用户平均接入效用，t表示迭代次数；Among them, S represents the set of all access networks, x_i (t) represents the proportion of users accessing network i in the t-th iteration to all users, Π_i (t) represents the number of users accessing network i in the t-th iteration user access utility, Represents the average access utility of all users accessing the network at the t-th iteration, and t represents the number of iterations;步骤ii)、对于各接入网络，将该接入网络中用户占所有用户的比例的稳定解乘以所有用户的数量得到该接入网络达到负载均衡的用户数。Step ii), for each access network, multiply the stable solution of the ratio of users in the access network to all users by the number of all users to obtain the number of users in the access network that achieves load balance.8.根据权利要求7所述的方法，其中，根据下式计算所有接入网络的用户平均接入效用：8. The method according to claim 7, wherein the average access utility of all users accessing the network is calculated according to the following formula:$\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}<span><span>=</span>=</span>\underset{\mathrm{<span><span>i</span>i</span>}<span><span>\&Element;</span>\&Element;</span>\mathrm{<span><span>S</span>S</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}{\mathrm{<span><span>x</span>x</span>}}_{\mathrm{<span><span>i</span>i</span>}}{\mathrm{<span><span>\&Pi;</span>\&Pi;</span>}}_{\mathrm{<span><span>i</span>i</span>}}$其中，S表示所有接入网络的集合，Π_i表示接入网络i的用户接入效用，x_i表示接入网络i中用户占所有用户的比例。Among them, S represents the set of all access networks,_Πi represents the user access utility of access network i, and_xi represents the proportion of users in access network i to all users.9.根据权利要求6所述的方法，其中，步骤B)包括对于各接入网络执行如下操作：9. The method according to claim 6, wherein step B) includes performing the following operations for each access network:步骤B1)、该接入网络统计接入用户数；Step B1), the access network counts the number of access users;步骤B2)、如果接入用户数小于该接入网络达到负载均衡的用户数并且当前网络偏移值小于预定阈值，则该接入网络将当前网络偏移值增加预定值并向用户广播当前网络偏移值，返回步骤B1)；其中，用户根据接入网络的当前网络偏移值和参考信号接收功率选择接入网络。Step B2), if the number of access users is less than the number of users that the access network achieves load balancing and the current network offset value is less than a predetermined threshold, the access network increases the current network offset value by a predetermined value and broadcasts the current network offset value to users. offset value, return to step B1); wherein, the user selects an access network according to the current network offset value of the access network and the received power of the reference signal.10.根据权利要求9所述的方法，还包括：10. The method of claim 9, further comprising:对于各接入网络，将该接入网络的当前网络偏移值减去所有异构接入网络的当前网络偏移值中的最小值作为该接入网络的当前网络偏移值。For each access network, the current network offset value of the access network minus the minimum value of the current network offset values of all heterogeneous access networks is used as the current network offset value of the access network.11.根据权利要求10所述的方法，还包括：11. The method of claim 10, further comprising:步骤4)、在经历预定时段后，返回步骤1)。Step 4), after a predetermined period of time, return to step 1).12.根据权利要求6所述的方法，其中，在步骤B)中，用户根据接入网络的当前网络偏移值和参考信号接收功率选择接入网络包括：12. The method according to claim 6, wherein, in step B), the user selects the access network according to the current network offset value and the reference signal received power of the access network comprising:用户选择该用户允许接入网络中当前网络偏移值与参考信号接收功率之和最大的接入网络。The user selects the access network with the largest sum of the current network offset value and the reference signal received power among the access networks allowed by the user.13.根据权利要求1或2所述的方法，其中，在步骤1)之前还包括：13. The method according to claim 1 or 2, wherein, before step 1), further comprising:初始化各接入网络的当前网络偏移值；Initialize the current network offset value of each access network;各接入网络向用户广播当前网络偏移值；其中，用户根据接入网络的当前网络偏移值和参考信号接收功率选择接入网络。Each access network broadcasts the current network offset value to the user; wherein, the user selects an access network according to the current network offset value of the access network and the received power of the reference signal.